What Is Electrical Activity in the Brain?

Electrical activity in the brain is the collective flow of charged particles, primarily sodium and potassium ions, across the membranes of billions of nerve cells. Every thought, sensation, movement, and memory involves neurons generating tiny voltage shifts that propagate through circuits and synchronize into larger rhythmic patterns detectable from the scalp. These signals span scales from a single ion channel opening for a few milliseconds to whole-brain oscillations that coordinate perception and consciousness, and understanding them has become central to treating neurological disease, building brain-computer interfaces, and deciphering how the mind works.

How a Single Neuron Generates Electricity

A neuron at rest is not electrically neutral. Its interior sits at roughly negative 60 to negative 70 millivolts relative to the outside, a difference maintained by protein pumps that continuously shuttle sodium ions out and potassium ions in. This resting voltage is sometimes called the membrane potential, and it exists because the cell membrane is selectively leaky: potassium seeps out more easily than sodium seeps in, leaving a net negative charge inside. The sodium-potassium pump is the engine behind this imbalance, and a coherent account of the resting potential can be built entirely from the pump’s transport currents, the membrane’s permeability to each ion, and the ion concentrations in the fluid surrounding the cell.¹

When a neuron receives enough input, its membrane voltage swings rapidly from negative to positive and back again in about one to two milliseconds. This spike, called an action potential, is the fundamental unit of fast communication in the nervous system. It works through a chain reaction: voltage-sensitive sodium channels snap open, sodium rushes in, the membrane voltage surges positive, and then potassium channels open to let potassium rush out and restore the negative resting state. In the cortex, fast potassium channels activated during an action potential have been measured opening at a threshold around negative 20 millivolts and staying active for roughly six milliseconds at resting potential.² The whole event is brief and stereotyped: a neuron either fires a full spike or it doesn’t, and the spike always looks about the same size. Information is carried not in the spike’s amplitude but in how often and when spikes occur.

Synaptic Signals and the Balance of Excitation and Inhibition

Action potentials travel down a neuron’s axon to its terminals, where they trigger the release of chemical messengers called neurotransmitters into the tiny gap between cells. These chemicals land on the next neuron and open its own ion channels, producing smaller, graded voltage changes. If the effect pushes the receiving cell’s voltage toward firing threshold, it’s excitatory; if it pushes the voltage further from threshold, it’s inhibitory. In the visual cortex, for instance, visually evoked inhibition predominantly works by hyperpolarizing the membrane, pulling the voltage in the negative direction and making the neuron less likely to fire.³

This balance between excitation and inhibition is not a one-time setting. It shifts constantly, and the brain can recalibrate it over longer timescales. When hippocampal pathways are stimulated intensely, the resulting strengthening of excitatory connections is greater than the strengthening of inhibitory inputs onto the same cells, tipping the balance toward easier firing.⁴ This kind of plasticity is thought to underlie learning and memory formation. It also hints at something important: the brain’s electrical activity is not just a readout of what’s happening. It actively reshapes the circuits that produce it.

From Single Cells to Brain Waves

One neuron’s action potential is far too small to detect from outside the skull. What an EEG picks up is the summed activity of thousands or millions of neurons firing in loose coordination. When large populations of cortical pyramidal neurons receive synaptic input at roughly the same time, their individual voltage changes add up to create a local field potential, a fluctuating electrical signal detectable with electrodes placed nearby. In the cortex, this signal is thought to stem mainly from correlated synaptic input to populations of geometrically aligned neurons.⁵

These local field potentials organize into rhythmic oscillations, commonly called brain waves. Different frequency bands are associated with different brain states and cognitive demands. Delta waves (below about 4 Hz) dominate deep sleep. Theta waves (4–8 Hz) appear during memory encoding and navigation. Alpha waves (8–14 Hz) are prominent when you close your eyes or relax, and their amplitude drops when you start looking around or attending to visual input. Beta (14–30 Hz) and gamma (30–70 Hz and above) oscillations become important during active thinking, perception, and working memory. Research argues that these selectively distributed oscillatory systems act as resonant communication networks linking large populations of neurons, playing a major role in memory and integrative brain functions.⁶

An increasingly influential idea is that no single frequency band works alone. Alpha oscillations were long dismissed as mere “idling,” the brain’s screensaver. But recent work has given alpha a more active role: its phase dynamics appear directly involved in attention and consciousness. One proposal is that simultaneous alpha, beta, and gamma oscillations are all required for unified cognitive operations, and that cross-frequency synchrony between them coordinates the selection and maintenance of mental representations during perception and working memory.⁷

Measuring Brain Electricity from Outside the Head

Electroencephalography, or EEG, has been the workhorse of brain electrical measurement since the 1920s. Electrodes on the scalp pick up voltage fluctuations generated in the cortex below. The signal is cheap, fast (millisecond resolution), and portable, which is why EEG remains central in clinical neurology, sleep medicine, and brain-computer interfaces. But it has a significant limitation: the skull acts as a low-conductivity barrier that blurs and smears the electrical signals passing through it. Features like skull thickness, holes, or lesions can distort the pattern enough to cause source localization errors of up to 15 millimeters.⁸ Volume conduction also creates spurious connectivity: a single cortical source can project onto several scalp electrodes simultaneously, making separate brain regions appear to be communicating when they are not.⁹

Magnetoencephalography, or MEG, sidesteps much of this problem. Instead of measuring electrical voltages, MEG detects the tiny magnetic fields produced by the same neuronal currents. Because magnetic fields pass through the skull without distortion, MEG offers better spatial resolution than EEG.¹⁰ A unipolar MEG detector “sees” a smaller patch of cortex than a comparable EEG electrode, precisely because EEG signals are smeared by the skull while MEG signals are not.¹¹ The trade-off is practical: MEG machines require supercooled sensors housed in large, expensive, magnetically shielded rooms, so the technique is mostly confined to research hospitals and specialized labs.

What fMRI Actually Measures (and What It Misses)

Functional MRI, or fMRI, is probably the brain-imaging technique most people have heard of, but it does not directly measure electrical activity. It tracks changes in blood oxygenation, working on the principle that active brain regions consume more oxygen and trigger increased local blood flow. The resulting signal, called BOLD (blood-oxygen-level-dependent), is an indirect proxy. A widely accepted model holds that the BOLD signal reflects synaptic activity in the form of local field potentials rather than the firing rate of individual neurons.¹²

That model works reasonably well for sensory cortex, but the relationship can break down. Under deep anesthesia, for example, the BOLD signal in frontal, parietal, and temporal regions increases after bursts of electrical activity and decreases during electrical suppression, as you’d expect. But in the occipital lobe, the opposite happens: BOLD goes down when electrical activity goes up.¹³ These dissociations are a reminder that fMRI maps of brain activation are interpretations of a metabolic signal, not direct recordings of the electrical events that actually carry information.

The Electrical Landscape of Sleep

Sleep offers some of the most dramatic shifts in brain electrical patterns. During slow-wave sleep, large, synchronized waves of activity sweep across cortical networks. Neurons alternate between “on” periods of vigorous firing and “off” periods of near-silence. These slow oscillations are accompanied by faster bursts called sleep spindles, and when spindles nest within the upstate of a slow oscillation, the combination is thought to support memory consolidation and synaptic plasticity.¹⁴

A striking finding from depth electrode recordings in human patients is that these slow waves are not as global as they appear on scalp EEG. On and off periods of spiking activity typically involve a limited number of brain regions rather than the entire cortex at once, and sleep spindles are mostly local events too. Slow waves do show a tendency to propagate from prefrontal cortex toward the medial temporal lobe and hippocampus, which may be one mechanism by which memories encoded during the day are reactivated and consolidated during sleep.¹⁵

When Electrical Activity Goes Wrong

Epileptic seizures are perhaps the most vivid example of disordered brain electricity. A seizure occurs when the normal balance of excitation and inhibition breaks down and large groups of neurons begin firing in abnormal synchrony. In temporal lobe epilepsy, genes, developmental mechanisms, and neuronal plasticity all contribute to creating a state of underlying hyperexcitability that can eventually erupt into seizure activity.¹⁶ The EEG is indispensable here clinically: characteristic spikes and sharp waves on an EEG recording can identify the seizure focus and help guide treatment, including whether surgical removal of the focus might be an option.

Anesthesia represents a different kind of electrical disruption, one that is intentional. General anesthetics progressively suppress cortical activity, shifting EEG patterns from the fast, desynchronized rhythms of wakefulness toward slow, high-amplitude oscillations and eventually burst-suppression, where brief bursts of activity alternate with flat silence. Monitors used in the operating room translate these complex electrical patterns into simple numerical indices to track anesthetic depth.¹⁷

Manipulating Brain Electrical Activity

Understanding brain electricity has opened the door to deliberately modifying it for therapeutic purposes. Deep brain stimulation, or DBS, delivers electrical pulses through electrodes implanted in specific brain structures. It is used for Parkinson’s disease, essential tremor, and some forms of dystonia and depression. Rather than working through a single mechanism, DBS appears to act via several routes simultaneously: local effects on neurons near the electrode, network-wide changes in oscillatory patterns, neurochemical shifts, and possibly even long-term synaptic plasticity and neuroprotection.¹⁸

Non-invasive alternatives work from outside the skull. Transcranial direct current stimulation, or tDCS, passes a weak current through scalp electrodes to shift cortical excitability. The direction of the current matters: anodal stimulation (positive electrode over the target area) depolarizes neurons and makes them more excitable, while cathodal stimulation hyperpolarizes them and dampens activity.¹⁹ Newer stimulation protocols have extended these effects well beyond the period of stimulation itself, opening the door to clinical applications in rehabilitation, depression, and chronic pain.²⁰

Brain-Computer Interfaces and Neural Decoding

Brain-computer interfaces, or BCIs, take the idea of reading brain electrical activity a step further: they translate neural signals into commands for external devices. Modern microelectrode arrays implanted in the cortex sample voltages at several thousand times per second, designed to record action potentials from individual neurons. An alternative approach uses local field potentials, acquired by filtering the raw voltage signal in lower frequency ranges (roughly 0.1 to 200 Hz). These signals reflect the spatially averaged electrical fields of groups of neurons and carry information about what a person intends to do.²¹ BCI research has demonstrated that people with paralysis can control robotic arms, cursor movement, and even typing interfaces using decoded brain electrical activity, though the technology remains largely in the research stage for most applications.

The Energy Bill for Brain Electricity

All this electrical signaling comes at a steep metabolic cost. The brain accounts for roughly two percent of body weight but consumes around twenty percent of the body’s resting energy. Much of that energy goes to the sodium-potassium pump, which works continuously to restore ion gradients after every synaptic event and action potential.²² Updated energy budgets for the rat cortex estimate that the processes mediating synaptic transmission, including neurotransmitter packaging and postsynaptic receptor activation, consume about 55 percent of the total energy spent on electrical signaling, making synapses by far the most expensive component.²³ This helps explain why the brain is so vulnerable to interruptions in blood supply: even a few minutes without oxygen depletes the ATP needed to run those pumps, and without them, the carefully maintained ion gradients collapse.

Electrical Activity Beyond Neurons

Neurons get most of the attention, but they are not the only brain cells with important electrical properties. Astrocytes, the most abundant type of glial cell, play a crucial supporting role in managing potassium ions that accumulate in the extracellular space after neuronal firing. Traditionally, astrocytes were thought to mop up excess potassium by physically moving it through gap junctions connecting neighboring astrocytes, a process called spatial buffering. But recent work suggests the picture is different: gap junctions may contribute to potassium clearance mainly by keeping the astrocyte membrane potential hyperpolarized, which sustains the driving force for potassium uptake rather than shuttling it physically across long distances.²⁴ Because extracellular potassium levels directly affect how excitable neurons are, astrocyte dysfunction can shift the brain’s electrical balance.

Electrical signals also matter long before neurons exist. During early brain development, endogenous electric fields orient cell division, direct the migration of newborn neurons, and establish cell polarity. Applied electric fields can orient hippocampal cell division and direct neuronal migration in culture, demonstrating that electrical cues are a fundamental guidance system for the developing brain.²⁵ More broadly, endogenous distributions of membrane potentials produced by ion channels and gap junctions are present across all tissues and play essential roles in pattern formation and morphogenesis, not just in the nervous system.²⁶

Ephaptic Coupling and Field Effects

Most discussions of brain communication focus on synapses, the chemical junctions between neurons. But there is another channel: ephaptic coupling, in which the extracellular electric fields generated by active neurons directly influence the membrane voltage of their neighbors, no synapse required. These effects are small under normal conditions, producing somatic membrane potential changes of less than 0.5 millivolts. Yet despite their tiny size, endogenous extracellular fields can strongly entrain action potential timing, particularly for slow fluctuations below about 8 Hz.²⁷ This means that the aggregate electrical field created by normal brain activity feeds back on the very neurons producing it, a subtle feedback loop that may help synchronize slow oscillations during sleep and other states without any wiring diagram to explain it.

How Electrical Excitability Evolved

It’s tempting to think of electrical signaling as something nervous systems invented, but the raw ingredients are far older. Voltage-gated ion channels, the proteins that make action potentials possible, predate the evolution of neurons entirely. Single-celled organisms and simple multicellular animals without nervous systems still use ion channels to adjust their behavior. A large-scale analysis of voltage-gated ion channel genes across 623 organisms found that the number of channels an organism has does not predict nervous system complexity. Instead, what distinguishes complex nervous systems is the diversity of channel types and their restricted expression in specific cell types.²⁸ In other words, neurons did not invent bioelectricity. They inherited it, diversified the toolkit, and parceled it out into specialized cell populations. Multiple animal lineages appear to have converged on this strategy independently, which suggests that the brain’s electrical activity is not one evolutionary accident but a solution that natural selection has discovered repeatedly.